Nuclear Forensics Analysis with Missing and Uncertain Data

Journal Article · Mon Oct 05 00:00:00 EDT 2015 · Journal of Radioanalytical and Nuclear Chemistry

DOI:https://doi.org/10.1007/s10967-015-4458-x· OSTI ID:1286883

Langan, Roisin T. ^[1]; Archibald, Richard K. ^[1]; Lamberti, Vincent ^[2]

---

1. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
2. Y-12 National Security Complex, Oak Ridge, TN (United States)

We have applied a new imputation-based method for analyzing incomplete data, called Monte Carlo Bayesian Database Generation (MCBDG), to the Spent Fuel Isotopic Composition (SFCOMPO) database. About 60% of the entries are absent for SFCOMPO. The method estimates missing values of a property from a probability distribution created from the existing data for the property, and then generates multiple instances of the completed database for training a machine learning algorithm. Uncertainty in the data is represented by an empirical or an assumed error distribution. The method makes few assumptions about the underlying data, and compares favorably against results obtained by replacing missing information with constant values.

• Research Organization: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
• Sponsoring Organization: USDOE
• Grant/Contract Number: AC05-00OR22725
• OSTI ID: 1286883
• Journal Information: Journal of Radioanalytical and Nuclear Chemistry, Vol. 1, Issue 1; ISSN 0236-5731
• Publisher: Springer
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING
Nuclear forensics
Missing data
Machine learning
Bayesian methods
Monte Carlo methods
Spent fuel isotopic composition (SFCOMPO) database